Supplementary Material for: Systematic Review of Observational Studies with Dose-Response Meta-Analysis between Folate Intake and Status Biomarkers in Adults and the Elderly

<b><i>Background:</i></b><i></i> Dietary reference values for folate intake vary widely across Europe. <b><i>Methods:</i></b> MEDLINE and Embase through November 2016 were searched for data on the association between folate intake and biomarkers (serum/plasma folate, red blood cell [RBC] folate, plasma homocysteine) from observational studies in healthy adults and elderly. The regression coefficient of biomarkers on intake (β) was extracted from each study, and the overall and stratified pooled β and SE (β) were obtained by random effects meta-analysis on a double log scale. These dose-response estimates may be used to derive folate intake reference values. <b><i>Results:</i></b> For every doubling in folate intake, the changes in serum/plasma folate, RBC folate and plasma homocysteine were +22, +21, and –16% respectively. The overall pooled regression coefficients were β = 0.29 (95% CI 0.21–0.37) for serum/plasma folate (26 estimates from 17 studies), β = 0.28 (95% CI 0.21–0.36) for RBC (13 estimates from 11 studies), and β = –0.21 (95% CI –0.31 to –0.11) for plasma homocysteine (10 estimates from 6 studies). <b><i>Conclusion:</i></b> These estimates along with those from randomized controlled trials can be used for underpinning dietary recommendations for folate in adults and elderly.